One falls slower in a mine shaft than in free air. This is due to collisions with the walls. How should one model the terminal velocity in the presence of such collisions?
Have a look at: http://iopscience.iop.org/0295-5075/60/2/220
A similar situation is described in: http://arxiv.org/abs/cond-mat/9904139
Terminal velocity is terminal velocity, regardless of collisions (Edit: I should clarify this... Terminal velocity is defined as the steady state velocity that you reach when a velocity dependent resistive force exactly cancels the force driving the motion [usually gravity]. Occasionally you'll hit a wall and velocity will be lost deforming the wall (or your body!) These collisions won't reduce the terminal velocity, but they will definitely reduce the average velocity and possibly prevent the falling object from ever even reaching terminal velocity.
As a first pass model, for fun, I'd suggest free fall in gravity, with the usual velocity dependent damping (air resistance) and a stochastic term that reduces the velocity according to a Poisson distribution.
This of course won't change the terminal velocity (which you'd get as the steady state in the collision-less limit), but it would give you a better estimate of the average velocity.